Mutation rate variation in multicellular eukaryotes: causes and consequences

Key Points Basic knowledge about the rate and range of mutation is central to our understanding of numerous evolutionary processes that include maintaining sexual reproduction and rates of molecular evolution. Although mutation rates are known to vary among species, little is known about the forces that underlie this variation at an empirical level, particularly in multicellular eukaryotes. The theoretical framework for mutational variation is based primarily on the 'cost of fidelity' and 'modifier allele' theories. The former argues that the mutation rate does not evolve to zero, despite the much greater frequency of deleterious mutations, because there is an opposing metabolic cost. The latter models mutational variation as the interaction between a mutator locus that affects the mutation rate for a fitness locus (or loci). Natural selection can potentially modulate the mutation rate through four main points of control: DNA replication fidelity, mutagen exposure, DNA repair efficiencies and the buffering of mutational effects. Molecular mutation rates are generally estimated through three approaches: gene-specific methods, mutation-accumulation lines and pedigrees. Although most empirical mutational knowledge derives from the first method, the other two probably provide more accurate estimates. Per-nucleotide mutation rates that are on the order of ∼10 −8 per generation are observed in Caenorhabditis elegans and Drosophila melanogaster mutation-accumulation experiments. Studies on the differences in mutation rates within and between genomic systems have the potential to provide a framework for understanding natural mutational variation. Animal mitochondrial genomes experience substitution rates that are much greater than those of nuclear genomes, whereas the situation is reversed for most plant species. Rate variation is also observed across different nuclear genomic regions within a species and might be affected by various forces, including base composition, recombination rate, gene expression, gene density and DNA repair domains. Estimates of the deleterious genomic mutation rate ( U ) are available for a variety of species that derive from fitness assay data and nucleotide substitution rates. Although according to these approaches U varies considerably across groups, the current evidence suggests that this parameter is rarely much less than one in multicellular eukaryotes and that there is as much variation within major lineages as between taxa. The r